AI and CMS image generated by Canva

How AI is revolutionising the CMS landscape

Luminary MD Adam Griffith reviews how some of the leading Content Management Systems are incorporating AI into their offerings.

Adam Griffith

03 March 2025

14 minute read

The meteoric trajectory of the uptake of AI over the last couple of years is well documented. From drug discovery, predictive wearables, and medical imaging in healthcare, to predictive maintenance and industrial robots in manufacturing (just to name a few). 

But what about the world of content management systems (CMSs) and digital experience platforms (DXPs)? How is AI changing the face of content management as we know it?

As someone who lives and breathes digital, tech, and software in general, I've been very closely monitoring the AI and machine learning space over the past few years, and while ChatGPT was only released in November of 2022, the application of AI to the CMS/DXP space has been touted for quite some time. Machine learning (ML), natural language processing (NLP) and pattern recognition have been successfully applied for many years now in automated personalisation, sentiment analysis, profanity detection, and automated image tagging. 

However, the release and widespread use of large language models (LLMs) from the likes of OpenAI, Anthropic, and Google has accelerated the application of AI to a broader range of CMS and DXP functions. 

But, why?

With any new tech or feature, we always start with the ‘why?’ and hone in on the benefit or value being promised. In the case of AI and content management systems, it comes down to two key benefits:

Productivity – this is all about being more efficient. Saving your organisation money and time. Features like summarising long articles to provide introductions or meta descriptions, auto-tagging images, and creating titles automatically from a blog post, are all about making the content editor more productive. 

Quality – this is the flip-side to productivity, and is all about delivering more impactful content and digital experiences, with the overarching goal of growth - growth in conversions, donations, members, sales, or whatever it is that your organisation essentially ‘sells’. Features like improving or simplifying language, or changing the tone of a selection of text to align it better with your brand tone of voice, are all about delivering content of a high quality. 

The AI-CMS landscape 

AI is being used in a variety of ways to enhance CMSs and DXPs. Some of the most common use cases include:

Content generation – for example, product descriptions, articles, headlines, summaries and social media posts. AI can also be used to suggest relevant keywords and tags.

Content personalisation – tapping into customer data, AI can be used to personalise content for individual users, or recommend relevant content, based on their profile and past behaviour. 

Search optimisation – AI can be used to optimise content for search engines, including tasks such as keyword research, meta description generation, and image optimisation. 

Workflow automation – including automation of tasks like scheduling content for publishing, tagging and categorisation of content, and routing content to the appropriate personnel for approval.

Predictive analytics – for example, analysing customer behaviour to identify patterns and predict which products individuals are most likely to purchase, highlight which customers are most likely to churn, and which products are most likely to be purchased together. 

How specific platforms are utilising AI

So, what are the leading enterprise CMS and DXP vendors offering when it comes to AI features?

Firstly, a disclaimer. As I type this, AI development is moving faster than a pre-election pollie’s promises. What’s new and shiny today is likely to be de rigueur tomorrow. So whilst the best endeavours have been made here to bring you the most up-to-date information, we’ve also provided links to the relevant pages on the platforms’ own websites for you to check out their latest AI related releases. 

Want to jump ahead to read about a specific platform? 

Kontent.ai

With ‘ai’ in its brand name, you would expect AI to feature prominently in this enterprise-focused headless CMS. And it absolutely does. Touting itself as the first headless CMS with native AI features, Kontent.ai boasts a long list of in-built AI-driven functionality to hit both the productivity and quality benefits. These features include:

Content creation – Kontent.ai allows the editor to create a targeted title from a completed article, summarise a selection of text, and generate entire paragraphs from a simple bullet point outline. 

Content refinement – the CMS also allows editors to refine existing content by fixing spelling and grammar, making content shorter or longer, simplifying complex language, changing the tone of a selection of text (e.g. to be more professional or friendly), and a rather vague option to simply ‘Improve writing’ (as this post was written in Kontent.ai, this paragraph has officially been ‘improved’ 😉). 

Content review – lets you automate your first round of reviews using your own content review guidelines.  

Semantic search – semantic search helps users land on the right content by connecting the dots between search terms, their context, related concepts, and the user’s intent, rather than simply matching exact keywords in results.

Translations – this feature uses AI to do a first parse translation, before your content editors add the final human touch.

Asset descriptions – generates search optimised image descriptions.

Smart taxonomy and linked item suggestions – the CMS can use AI to read through texts and suggest the right taxonomy terms, as well as suggesting relevant linked items. 

It’s important to note here that these features are native to the platform and don’t require any integration with a third party service. Kontent.ai also promises that your data and content is secure and will not be used to train OpenAI or other large language models (LLMs) beyond the application to your own collection of content. This effectively means the AI will get to know your tone of voice and style over time, and hopefully be able to provide more refined and on-brand outputs for your team. 

Check out Kontent.ai’s latest AI developments here.

To get a good visual on Kontent.ai's AI features, watch this demo video.

Optimizely

The Optimizely suite has been utilising machine learning and pattern recognition longer than most of the vendors in this list. ‘Optimizely Opal’, the brand’s embedded AI assistant for marketing, promises to accelerate every phase of the digital experience lifecycle, from ideating and creating content, to anticipating and automating administrative tasks. Opal is informed by AI agents that are guided by your unique company data, brand guidelines and instructions. It applies throughout all of the workflows across the Optimizely One platform. 

Key features of Optimizely Opal include:

Campaign brief ideation – Optimizely Opal references past performance data to suggest topics and define objectives, allowing for rapid campaign development. Opal generates campaign-specific assets, images, social posts and more, using context from the brief. It also generates new copy within the Visual Builder, helping you brainstorm new variations, ideate on new CTAs and fine-tune your positioning.

Brief-based task generation – Opal anticipates tasks to be completed that are unique to each campaign, freeing marketers to focus on creative execution rather than administration. This includes suggesting audience segments for targeting, providing an instant snapshot of any customer and applying user-level attributes, profiles and real-time behaviour to personalise the experience for every visitor.

Recommendations – including Content Intelligence, Content Recommendations, and Product Recommendations, this feature-set was extended via the acquisition of Idio in 2019. It provides editors and marketers with the ability to analyse their content landscape, pinpoint topic areas that are performing well (or underperforming) and focus in on the content that needs to be created. Once content has been created and is in play, it can then dynamically deliver the most relevant content (including product-based content) to a particular user based on machine learning driven by content performance and user behaviour. This applies to both web and email channels. 

Check out Optimizely’s latest AI developments here.

Optimizely Opal

Sitecore

Sitecore’s native AI offering is called ‘Sitecore Stream’. Officially launched at the Sitecore Symposium in October 2024, Sitecore Stream encapsulates all of the composable DXP’s native AI functionality. Built on Microsoft Azure OpenAI Service, Sitecore Stream offers ‘AI-enhanced workflows, generative copilots, and brand-aware AI capabilities’ across its suite of products, including XM Cloud, Content Hub, and Experience Platform (XP). 

The key features of Sitecore Stream are:

AI-enhanced workflows – can be used to automate simple tasks like SEO tagging or A/B testing a call to action, through to more complex executions such as generating personalised content or building a web page based on a brief.

Generative copilots - intuitive conversation-like chatbots designed to assist with brand, brief, content, and experience creation and optimisation. The ‘brand copilot’ helps manage everything from brand kits, to content assets and brand guidelines, while the ‘content copilot’ helps create on-brand content – from headlines to campaigns – for different audiences while maintaining brand consistency and supporting multi-author collaboration. The ‘experience copilot’ helps build elements like web pages, landing pages, and individual components like forms.   

Brand-aware AI – Sitecore Stream leverages assets such as brand guidelines, value statements, and visual identity kits to ensure that all AI-driven outputs are on-brand. This can apply to anything from colour schemes and messaging, through to key market trends. 

While Sitecore Stream has access to the brand’s knowledge repositories, this information is not used to train the underlying AI models, to prevent information leakage. It can also be configured to a company’s security protocols to ensure that only authorised users can access specific information. 

Check out Sitecore’s latest AI developments here.

Sitecore Stream

Xperience by Kentico

Xperience by Kentico features AIRA – the platform’s native Artificial Intelligence Recommendations and Assistance capability. AIRA combines AI models with administrator-authorised access to content and data stored in Xperience by Kentico. This enables AIRA to utilise relevant information to make decisions and recommendations. It can adapt written or visual content to meet the requirements of various channels, including websites, emails, social media or mobile apps.

It has the capacity to:

  • Ensure consistent brand tone of voice of content across channels
  • turn articles into newsletters
  • create image format variants for your website, social media and newsletters
  • optimise assets to meet the requirements of specific channels
  • automate image format conversion to ensure quality optimisation during mass file uploads
  • automatically add tags, alt texts and image descriptions based on visual content
  • automatically crop images, identifying image focus points so that the cropping preserves key visuals
  • assign tags for content items in the content hub, avoiding the need for manual input
  • provide guidance on using Xperience by Kentico.

In combination with the AIRA Companion App, marketers and content editors can also access their data stored across the DXP, even from a mobile device. 

To maintain data privacy and security, AIRA uses Azure OpenAI, which does not use your data to retrain models – all of the data remains within your Kentico instance.

Check out Kentico’s latest AI developments here.

Contentful 

Contentful leans heavily on its suggested marketplace of AI-powered integrations and plugins. It does, however, offer a number of integrated AI features, many of which came with its recent acquisition of Ninetailed, including:

AI suggestions – enables marketers to refine audiences, experiments and personalised experiences using Ninetailed by Contentful.

Variant generation – utilises generative AI to leverage your content, customer data and Ninetailed’s Knowledge Base.

Smart traffic distribution – unlike traditional experimentation methods, which divide resources equally regardless of outcome, Contentful use of Multi-Armed Bandit continuously adjusts its strategy to favour the most effective options.

AI content type generator – designs and builds content models.

AI content generator – generates on-brand content at scale, with the ability to translate what you create into different languages.

AI image generator – leverages OpenAI to generate images and then deploy them on the Contentful platform.

Check out Contentful’s latest AI developments here.

Contentful AI

Contentstack

In the same vein as Contentful, Contentstack describes itself as a ‘composable CMS built to integrate with AI-enabled content management tools’. 

Contentstack’s native AI features include:

AI Assistant – this allows you to use AI for real-time translations, summaries and idea generation directly within the authoring experience. You can also save custom actions for reuse and improve your AI prompts by leveraging your platform content to provide context.

AI Connectors – allows you to improve content operations using automated workflows.

Brand Kit – utilises Contentstack’s Knowledge Vaults – a repository of brand assets such as documents, customer insights, and platform context – to centralise key brand assets and voice profiles to set specific rules for tone, style and language across all AI-driven content production.

Check out Contentstack’s latest AI developments here

A warning

Just as AI can help your teams be more productive and produce higher-quality content and digital experiences, it can also do the opposite... Your team can easily waste precious time messing around with all these new features and lose sight of what their goal is; that is, to create and deliver impactful content. AI also allows them to be lazy and rely on it to help create content, thus risking the creation of content that is off-brand, or even worse, plain wrong (if you haven't read about AI hallucinations, I'd strongly recommend you do before going any further.)

Leveraging the capabilities of AI will also be dependent on a change in content operations in most organisations. For example, content will need to be created and reviewed within the CMS. In our experience this rarely happens, especially if the CEO has to review content prior to publishing and is unwilling to engage with a CMS editing interface… 

Caution should also be exercised when considering any new AI tool in terms of data security. Wherever possible, you should prioritise tools offered by trusted vendors with robust privacy and security practices. For more on minimising the security risks associated with AI, check out our blog post on Regulating the use of generative AI in the workplace

What about my CMS/DXP? 

So you're stuck on a legacy version of your CMS and suffering from FOMO. How do you get your fair share of AI? Start by exploring your current CMS's upgrade path and whether the latest version incorporates this functionality, or is at least on the roadmap. If a solution isn't imminent, you can simply resort to copy-paste, by utilising ChatGPT, Gemini, or Claude in the browser. Another simple alternative is to utilise GrammarlyGO which essentially does a lot of the content-related AI functions described above and is free for up to 100 prompts per month. With its Chrome Extension, you can bypass the copy-paste step in most cases. 

Should AI be on my shopping list when selecting a new CMS or DXP? 

If you're currently looking for a new CMS or DXP, I'd recommend reading our in-depth CMS selection guide that includes a downloadable selection matrix. We haven't yet incorporated AI as a critical selection criterion. Why not? Put simply, it's going to very quickly become table stakes. That is, within a year or so, every leading CMS and DXP vendor will incorporate the AI features we've discussed in this article (and some more that are likely to be developed over the next little while). It's not that complex, as the vendors aren't building the large language models themselves, but are instead relying on the big players like OpenAI and Google to drive the true AI features. 

However, of the six consideration areas we cover in the CMS selection guide (Investment, Capability, Usability, Security, Scalability, and Reliability), the one that will differentiate CMS and DXP platforms moving forward in their application of AI, is usability. How well these vendors design and implement these AI features for content editors to create and deliver quality content, productively, could well make the difference in the selection process. 

But will AI take my job? 

This post wouldn't be complete without some comment on the impending doom predicted for the job market. Sure, the kind of AI I've referenced above is certainly going to take paid tasks away from knowledge workers, but it's the lower-value tasks and this means more time can be allocated to the higher-value creative, strategic, and more human tasks (that we all prefer anyway!). 

In short, if you're reading this post (and made it this far 🙏), you'll be fine. 


Luminary aims to make the human experience brighter – for our clients, for their customers, and for our team. 

As a digital agency, we have a responsibility to lead the way with innovative, valuable technology – not just the latest fad. We are committed to putting people first in our work with AI and we will continue to explore ways that AI can enhance the work we do. 

Learn more about our approach to AI , and how we use it continue to have genuinely human-centered experiences.


Want to tap into the expertise of an agency that’s been in operation since 1999?

Get in touch

Keep Reading

Want more? Here are some other blog posts you might be interested in.